[ADMB Users] MVN Example?

[Chris Gast]

After changing line 1978 in df1b2fun.h to match the correct function header
on line 220 of  df1b2ludcmp.cpp, I got your syntax working successfully.

I'm not sure who specifically to address regarding making this correction,
but hopefully they're reading this thread already....?

Mollie: I'm interested in what complications you experienced with solve().
Is this related to some speed issues which I found noted on previous
admb-users posts?



Thanks Mollie (and Jim Ianelli for an off-list reply).

Chris




-----------------------------
Chris Gast
cmgast at gmail.com


On Wed, Mar 14, 2012 at 4:52 PM, Chris Gast <cmgast at gmail.com> wrote:

> Thanks, Mollie.  For some reason, ADMB can't seem to find the ln_det()
> function.
>
> test_mat.obj:test_mat.cpp:(.text+0x13af): undefined reference to
> `ln_det(df1b2matrix&)'
>
> I'll keep playing with it and see if I can figure it out.
>
>
> Chris
>
>
>
>
>
> -----------------------------
> Chris Gast
> cmgast at gmail.com
>
>
>
> On Wed, Mar 14, 2012 at 4:39 PM, Mollie Brooks <mbrooks at ufl.edu> wrote:
>
>> Hi Chris,
>> I'm doing an MVN model. My procedure section looks like this
>> PROCEDURE_SECTION
>> for(int i=1; i<=ncoeffs; i++)
>>  {
>> coeffs(i)=ucoeffs(i)*scale(i);
>> }
>>  nLL=0;
>> for(int p=1; p<=nponds; p++)
>> {
>> calc_predictions(p);
>>  calc_varcov(p);
>> dvar_vector diff=predictions(p)-obs(p);
>> dvar_matrix tmp_varcov=varcov(p);
>>  nLL+=.5*(nobs(p)*log(2*M_PI)+ln_det(tmp_varcov)+
>> diff*solve2(tmp_varcov,diff));
>> }
>>
>> The first for loop scales all my coefficients to make it conditioned
>> better.
>> Then I have observations of a bunch of ponds and each pond is assumed to
>> be independent of the others.
>> I defined functions calc_predictions() and calc_varcov() to fill in those
>> objects; the code in these is pretty specific to my problem. The important
>> thing fir efficiency is that my covariates are in a design matrix created
>> using model.matrix() in R.
>>
>> Try using solve() where I have solve2(). If you have a problem with
>> solve, I'll explain the complications of why I used this version.
>>
>> I hope this helps.
>> cheers,
>> Mollie
>>
>> Mollie Brooks
>> Ph.D. Candidate
>> NSF IGERT Fellow
>> Biology Department
>> University of Florida
>> mbrooks at ufl.edu
>> http://people.biology.ufl.edu/mbrooks
>>
>>
>>
>>
>> On 14 Mar 2012, at 5:50 PM, Chris Gast wrote:
>>
>> Hello,
>>
>> Does anyone have some code to compute the log-likelihood for a
>> multivariate normal model that they're willing to share? I'm having trouble
>> getting matrix determinants and inverses to work.
>>
>> I found a link to a previous discussion, indicating such an example would
>> be available in Roa's article "A Likelihood-Based Model of Fish Growth
>> With Multiple Length Frequency Data" in JABES, but I'm having trouble
>> accessing the supplementary material.
>>
>> If I can get this working (a weighted LMM with autocorrelated residuals
>> and spatially-correlated random effects), I'd be happy to add it to the
>> examples list, since I can't seem to find a MVN likelihood example there
>> either.
>>
>> Thanks,
>>
>> Chris
>>
>>
>>
>>
>>
>>
>> -----------------------------
>> Chris Gast
>> cmgast at gmail.com
>>  _______________________________________________
>> Users mailing list
>> Users at admb-project.org
>> http://lists.admb-project.org/mailman/listinfo/users
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.admb-project.org/pipermail/users/attachments/20120314/633d9bce/attachment.html>